[Bug 2183] Thread migration is broken on SMP

bugzilla-daemon at rtems.org bugzilla-daemon at rtems.org
Tue Jun 24 14:17:37 UTC 2014


--- Comment #5 from Joel Sherrill <joel.sherrill at oarcorp.com> 2014-06-24 09:17:37 CDT ---
(In reply to comment #4)
> (In reply to comment #3)
> > No. I don't think we have seen it looping infinitely. Just weird end results
> > where logically the scheduler wants to migrate a thread but it ends up with it
> > running on two processors. We are narrowing in on that and it sounds like a
> > completely different problem.
> Do you see this problem on a real RTEMS target or on the scheduler simulator?

Real hardware and grsim. Exactly the same.

> > 
> > I was concerned because the non-affinity scheduler avoids ever moving a thread
> > to another core but scheduler and affinity changes result in migration of
> > executing threads.
> The migrations work quite well in general except for the problem in this bug
> report, see for example smptests/smpmigration02.

OK. We will keep narrowing down on this problem.

Configure bugmail: https://www.rtems.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching all bug changes.

More information about the bugs mailing list